ComfyUI

ComfyUI is a visual AI workflow editor for creators who want control over generation and editing. Its node canvas lets you connect models and processing steps, inspect their settings and reuse the resulting workflows without writing code.

• Media generation: Build image, video, audio, 3D and text workflows with native support for models including Stable Diffusion, Flux, Qwen Image, Wan, LTX-Video, Hunyuan3D and Gemma.
• Editing and finishing: Combine inpainting, outpainting, masks, compositing, upscaling and frame interpolation within the same workflow.
• Reusable workflows: Start with community templates, group steps into reusable subgraphs, and use App Mode for a simpler view of a finished workflow.
• Application integration: Extend the editor with custom nodes or connect workflows to applications through its local API.

The local software is open source under GPL-3.0 and runs on Windows, macOS and Linux; the desktop app is available for Windows and macOS. The core can work fully offline. Optional paid API nodes send requests to external model services, including access to closed models such as Nano Banana and Seedance. Comfy Cloud is a hosted alternative, and the separate developer platform deploys workflows as production APIs.

Supported hardware includes NVIDIA, AMD and Intel GPUs. Model offloading, quantized models and memory management reduce memory demands, but hardware requirements depend on the models and workflow.